Well, the problem is in some ways similar to the problem encountered over the weekend. It seems the upload server is constipated. That doesn't stop access to the database pages (including the message boards and user stats), but it does leave SETI about as functional as the UN Security Council.

Well I got this after I uploaded a WU;
12/08/2008 06:17:01|SETI@home|Scheduler request failed: Couldn't connect to server
Yet I am sending this message, weird or what?

---

Thnak you is that why I am now getting
12/08/2008 08:13:20|SETI@home|Scheduler request failed: Failed sending data to the peer
after I got internet access ok, project may be down

---

Well I got this after I uploaded a WU;
12/08/2008 06:17:01|SETI@home|Scheduler request failed: Couldn't connect to server
Yet I am sending this message, weird or what?

The server being used for the forums/web pages, and the server that handles uploads/downloads/scheduler requests etc are not one & the same.
From memory they also use different network connections.

---

Grant
Darwin NT

Ah, children...
It would appear the log jam may have cleared.
Outbound network traffic 91Mb/s
Inbound network traffic 10Mb/s
Result creation rate 32.5/s

---

Grant
Darwin NT

Ah, children...
It would appear the log jam may have cleared.
Outbound network traffic 91Mb/s
Inbound network traffic 10Mb/s
Result creation rate 32.5/s

That looks like a blip.
Ready to send is at 0. As of 40 mins.
Result creation rate 1.2/sec.

CPDN & Rosetta are benefiting from this.

---

It would appear the log jam may have cleared.

Bugger!
Looks like it was just a minor brach, the dam is still intact.
:-/

---

Grant
Darwin NT

Not looking good. Something seems to have crashed (again).

Server status page numbers have flatlined for about 5 hours and cricket bits out are almost non existent.

---

Not looking good. Something seems to have crashed (again).

Server status page numbers have flatlined for about 5 hours and cricket bits out are almost non existent.

Server status page hasn't updated for over 10 hours now, hence the flatlining. Combined with Cricket showing almost no data, it looks like the staff will need to kick a few things when they get into the lab this morning.

                                                                Joe

Seti shines compared to the UNSC, considering what the ruskies are doing right now in georgia. But that is off topic.

I say turn off AP wu production now and clean up the ghost wu's. When normal seti production (non-ap) is stabilized again, then turn on only one ap splitter to gauge the effect on the overall system. That's my soap box and I'm going to stay on it.

Well, the problem is in some ways similar to the problem encountered over the weekend. It seems the upload server is constipated. That doesn't stop access to the database pages (including the message boards and user stats), but it does leave SETI about as functional as the UN Security Council.

Well I got this after I uploaded a WU;
12/08/2008 06:17:01|SETI@home|Scheduler request failed: Couldn't connect to server
Yet I am sending this message, weird or what?

Uhh.. I think it's time again for this one..



All people around, press this button to feel better..! ;-D
I pressed him few times.. many times.. hope he's not broken now.. ;-D


My 24/7 rig is nearly running dry.. I start to be nervous.. hope the server will send soon new work..

(BOINC update button ;-)

---

Hey, I just noticed the Cricket is chirping again!

Hey, I just noticed the Cricket is chirping again!

I got a few shortie WUs right before I noticed that chirping Cricket!

QS

. . . All's Well - that ends Well ;)

plenty of crunchin' here on mi box



. . . runs like hell to escape the baCkfirEs ;))

---

BOINC Wiki . . .

Science Status Page . . .

Hey, I just noticed the Cricket is chirping again!

I got a few shortie WUs right before I noticed that chirping Cricket!

QS

When the server stats stopped updating, 6 of the 9 mb_splitter processes were working on data within the 2 June through 20 June recorded time range. The Arecibo schedule shows that only the GALFACTS project was scheduled to use ALFA during that period, so a very high proportion of shorty work is likely from those recordings. ALFA of course might be running during some maintenance times, and the set up and tear down for GALFACTS observations may also generate some work which isn't Very High Angle Range shorties.

I'm still convinced that those many shorties are a major contributor to the server difficulties. If a normal mix requires 60 MBits/sec, a mix with many shorties can easily need 120 MBits/sec which simply isn't available. And those WUs are the same size as lower angle ranges so shorties greatly increase workunit storage requirements. The addition of AstroPulse may have been "the straw which broke the camel's back", but I think the servers would have been badly stressed even without that.

                                                                  Joe

Woo Hoo! Current result creation rate 32.82/sec [As of 13 Aug 2008 18:00:14 UTC]

We have lift off.

...and Cricket's green graph is showing 90+MB for a while now!

QS

---

Of course the stats are not generating again so BoincStats is showing nothing for the last couple of days.

---

I would just like to see things stabilize and see the optimizers be able to get the code to the point that credits would eqaulize between MB an AP.......
'Course, my MB credits are not what they used to be........

AP is the first seti@home app that does more than just waste time and electricity; It can actually do real science while it's looking for spooks. I say to heck with credit considerations and full speed ahead with AstroPulse!

A lot has been said lately about seti@home's shoestring budget. Well, AstroPulse is exactly the sort of respectable app to just possibly rectify that. Fact is, it's hard to get grants for chasing ghosts. Especially when you've been at it for years and have never even seen one, let alone caught one. It lacks credibility. The smart money doesn't go for it. But after we discover a few pulsars, and maybe even something really unexpected (fingers crossed)... I would imagine that both seti@home and Arecibo would get more respect.

Of course the stats are not generating again so BoincStats is showing nothing for the last couple of days.

Arkayn--please be patient. The stats will catch up

---

A lot has been said lately about seti@home's shoestring budget. Well, AstroPulse is exactly the sort of respectable app to just possibly rectify that. Fact is, it's hard to get grants for chasing ghosts. Especially when you've been at it for years and have never even seen one, let alone caught one. It lacks credibility. The smart money doesn't go for it. But after we discover a few pulsars, and maybe even something really unexpected (fingers crossed)... I would imagine that both seti@home and Arecibo would get more respect.

I agree most fully. So this is the reason the servers and the procedures at centralcommand should be as professionally operated as possible. Nobody is going to drop $1M for new servers. But a rock solid operational history that funding agents can 'trust' would go along way to justifying a post-doc and tech sized grant. That would be a start. The operations for the last week or so since turning on AP is disappointing, however.